Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

EM-Net: Efficient Channel and Frequency Learning with Mamba for 3D Medical Image Segmentation

About

Convolutional neural networks have primarily led 3D medical image segmentation but may be limited by small receptive fields. Transformer models excel in capturing global relationships through self-attention but are challenged by high computational costs at high resolutions. Recently, Mamba, a state space model, has emerged as an effective approach for sequential modeling. Inspired by its success, we introduce a novel Mamba-based 3D medical image segmentation model called EM-Net. It not only efficiently captures attentive interaction between regions by integrating and selecting channels, but also effectively utilizes frequency domain to harmonize the learning of features across varying scales, while accelerating training speed. Comprehensive experiments on two challenging multi-organ datasets with other state-of-the-art (SOTA) algorithms show that our method exhibits better segmentation accuracy while requiring nearly half the parameter size of SOTA models and 2x faster training speed.

Ao Chang, Jiajun Zeng, Ruobing Huang, Dong Ni• 2024

Related benchmarks

TaskDatasetResultRank
Abdominal multi-organ segmentationBTCV
Spleen88.34
58
Multi-organ SegmentationSynapse
Average DSC83.95
33
Brain Tumor SegmentationBraTS 2020
DSC (WT)90.81
27
Multi-organ SegmentationBTCV
Spl Score92.43
22
3D Brain Tumor SegmentationBraTS 2020
WT Dice90.81
11
Brain Tumor SegmentationBraTS 2023
Dice (WT)93.04
11
Showing 6 of 6 rows

Other info

Code

Follow for update